Current practice in radiotherapy struggles with the geometric uncertainties of treatment delivery.beam radiotherapy, such uncertainties arise from internal organ motiOn and setup errors, and, in brachytherapy these uncertainties are inherent to the invasive nature of the implant procedure. In either case, uncertainty can result in a geOgraphic miss of the targeted volume or, if margins are applied, a compromise in the prescription dose. Methods for improving therapeutic precision are clearly needed if radiotherapy is to be exploited to its fullest potential. In this proposal, a new imaging technology - flat-panel cone-beam CT (FPI-CBCT) - is evaluated for broad application in image-guided radiotherapy - allowing the therapist to acquire high quality, volumetric images during the therapy procedure. This approach improves therapeutic precision through imaging and feedback, providing guidance and assessment of the therapy procedure. FPI-CBCT is made possible by advances in two key technologies: cone-beam CT reconstruction methods, and the availability of flat-panel imagers (FPIs). In this proposal, the hypothesis that FPI-CBCT can satisfy the requirements for image-guidance in radiotherapy is tested through a systematic program of research. In Specific Aim 1, the physical factors affecting FPI-CBCT are quantified and the dependence of these factors on geometry and mode of operation is determined. In Specific Aim 2, a methodology for evaluating and predicting volumetric signal and noise transfer characteristics in FPI- CBCT is developed. Finally, Specific Aim 3 focuses the results of the first two aims to evaluate the imaging performance of FPI-CBCT for six clinically relevant image-guidance tasks. Successful completion of this program will identify an imaging solution that addresses the clinical need for image-guidance in radiation therapy.